A new product will be added to your cart each time you click the Product button.
If additional options exist (checkboxes) to further customize your product and you'd like to choose one of the them, please select the option before clicking the product button to add to your cart.
Click here to edit your cart.
All prices shown in USD | Click here to view our digital terms
ACTEX Learning Presents: ACTEX eLearning Webinar: Machine Learning in R for Actuaries
The world of machine learning is a dazzling, complicated place filled with great promise. A vast array of names, complicated techniques and strange, new software packages can make it feel overwhelming. It also feels pretty relevant. We've all heard the hype about data scientists' emergence as the “sexiest” profession of the 21st century. Where do actuaries fit in this emerging field?
Good news! The rise of so many diverse machine learning algorithms means that new tools have emerged to place them in a consistent framework. One such tool is the "caret" package for R. This package provides a common interface for over 200 different models. (We won't talk about all of them.) We'll walk through that package and a few others. Along the way, we will try to sift through the jargon and present tangible, practical examples that are pertinent to actuaries.
Learning Objectives:
- Understand how to adjust your data for modeling
- Complete an overview of the most popular machine learning techniques
- Learn how to train, evaluate and select from several different models
- Identify how to present results in a way that stakeholders will understand
Who Should Attend This Webinar?
This webinar is for any actuary who wants to expand their toolkit beyond linear models. Some familiarity with machine learning methods is useful, but we will make the material accessible to the novice as well.
We feel that participants will get the most out of the seminar if they do not try to go through the examples during the webinar itself. You will have access to the recording for 180 days, so you can work with the examples after the live event. Nevertheless, code examples will be provided before the seminar.
Here’s What Participants Said About Brian’s Last Webinar:
“I thought this presentation was PERFECT for someone like me … I think Brian explained the intention of the ggplot2 package in the way that Hadley Wickham would want it explained and it is an explanation that I had been missing prior to the presentation. A lot of lightbulbs went off in my head during this presentation. … This was an impactful presentation that will actually improve the efficiency and quality of my work (and I can't say that about too many presentations...).”
“As an entry-level candidate, I found that this seminar can be a huge benefit in my career search with respect to my ePortfolio. Having case studies or other data analyses with the plots I've learned today should make for some serious second looks into the artifacts themselves.”
“It was fantastic, I use R nowadays a lot, but wasn't aware of having such a good package for plotting. The presentation itself was also very good, tempo perfect, I hope I can listen to you once again.”
“Appreciated the step-by-step walk through from data to client-ready visualization.”
About Your Instructor
Brian A. Fannin, ACAS
Please take a moment to read Brian's story in the ACTEX Blog: "My Journey with R"
Brian is the founder and captain of PirateGrunt LLC, an actuarial, data and predictive modelling consultancy. Brian has held a number of positions at both primary and excess insurance companies, both in the US and overseas. He is the chair of the CAS open source software committee, with a focus on R. His principle areas of research are in stochastic reserving, predictive modelling and visualization of data. He is the author of the MRMR, raw, represtools and Imaginator R packages.
Important CPD Information:
Recorded ACTEX Webinars do not qualify for EA credit under the guidelines of the JBEA. ACTEX Webinars are strictly for educational purposes and cannot be submitted to attest for EA core or non-core credit.